APPENDIX
2.7 Empirical Tests of the Ricardian Model
BB∗=60W is the maximum quantity of wheat that the United Kingdom could produce at
the constant opportunity cost of PW/PC =2 (as in the right panel of Figure 2.2). Thus, 240W is the maximum combined total quantity of wheat that the United States and the United Kingdom could produce if both nations used all of their resources to produce wheat.
As a result, theSW(US+UK) curve is vertical at 240W.
Suppose that, with trade, the combined demand curve for cloth of the United States and the United Kingdom is DW(US+UK), as shown in the left panel of Figure 2.3.DW(US+UK) intersectsSW(US+UK)at pointE, determining the equilibrium quantity of 180W and the equi- librium relative price ofPW/PC =1 with trade (the same as in the left panel of Figure 2.2).
Note that, with trade, wheat is produced only in the United States, and the United States specializes completely in the production of wheat.
We can do the same for cloth. In the right panel of Figure 2.3,SC(UK+US)is the combined supply curve of cloth of the United Kingdom and the United States if both countries used all of their resources to produce only cloth. The United Kingdom can produce a maximum of 120C=0Bat the constantPC/PW =1/2and the United States can produce a maximum of another 120C=BBat the constantPC/PW =3/2(as in Figure 2.2).
Suppose that, with trade, the combined demand for cloth of the United Kingdom and the United States is DC(UK+US), as shown in the right panel of Figure 2.3. DC(UK+US) intersects SC(UK+US) at point E, determining the equilibrium quantity of 120C and the equilibrium-relative price of PC/PW =PW/PC =1 (the same as in the right panel of Figure 2.2). Note that, with trade, cloth is produced only in the United Kingdom, and the United Kingdom specializes completely in the production of cloth.
Finally, note that with complete specialization in production in both countries, the equilibrium-relative commodity price of each commodity is between the pretrade relative commodity price in each nation (see both panels of Figure 2.3). However, if in the left panel of Figure 2.3 DW(US+UK) were lower and intersected SW(US+UK) between points 0 andB on the horizontal portion ofSW(US+UK) at PW/PC =2/3, trade would take place at the pretrade relative commodity price of wheat ofPW/PC =2/3in the United States and the United Kingdom would receive all the gains from trade. This would occur if the United Kingdom were a small country that specialized completely in the production of cloth and the United States were larger and did not specialize completely in the production of wheat (see Problem 10, with answer at www.wiley.com/college/salvatore). This is known as the small-country case and shows the “importance of being unimportant.” This benefit, how- ever, is not without cost since the small nation (here, the United Kingdom) faces the risk of a possible future reduction in demand for the only commodity it produces.
2.7 Empirical Tests of the Ricardian Model
We now examine the results of empirical tests of the Ricardian trade model. We will see that if we allow for different labor productivities in various industries in different nations, the Ricardian trade model does a reasonably good job at explaining the pattern of trade.
The first such empirical test of the Ricardian trade model was conducted byMacDougall in 1951 and 1952, using labor productivity and export data for 25 industries in the United States and the United Kingdom for the year 1937.
Since wages were twice as high in the United States as in the United Kingdom, Mac- Dougall argued that costs of production would be lower in the United States in those industries where American labor was more than twice as productive as British labor. These
1 2 3 4 5 6
0.05 0.1 0.5 1.0 5.0
U.S. Exports U.K. Exports Output per U.S. Worker
Output per U.K. Worker
Woollens and Worsteds
Margarine
Beer Cotton Coke
Clothing
Rayon yarn Cement
Tin cans
Linoleum Cigarettes Hosiery
Rayon
cloth Footwear
Pig iron Machinery
Paper
Glass containers
Motor cars
Radios
FIGURE 2.4. Relative Labor Productivities and Comparative Advantage—United States and United Kingdom.
The figure shows a positive relationship between labor productivity and export shares for 20 industries in the United States and the United Kingdom, thus confirming the Ricardian trade model.
Source: Adapted from G. D. A. MacDougall, ‘‘British and American Exports: A Study Suggested by the Theory of Comparative Costs,’’Economic Journal , December 1951, p. 703.
would be the industries in which the United States had a comparative advantage with respect to the United Kingdom and in which it would undersell the United Kingdom in third markets (i.e., in the rest of the world). On the other hand, the United Kingdom would have a com- parative advantage and undersell the United States in those industries where the productivity of British labor was more than one-half the productivity of American labor.
In his test MacDougall excluded trade between the United States and the United Kingdom because tariffs varied widely from industry to industry, tending to offset the differences in labor productivity between the two nations. At the same time, both nations faced generally equal tariffs in third markets. The exclusion of trade between the United States and the United Kingdom did not bias the test because their exports to each other constituted less than 5 percent of their total exports.
Figure 2.4 summarizes MacDougall’s results. The vertical axis measures the ratio of output per U.S. worker to output per U.K. worker. The higher this ratio, the greater the relative productivity of U.S. labor. The horizontal axis measures the ratio of U.S. to U.K.
exports to third markets. The higher this ratio, the larger are U.S. exports in relation to U.K. exports to the rest of the world. Note that the scales are logarithmic (so that equal distances refer to equal percentage changes) rather than arithmetic (where equal distances would measure equal absolutechanges).
The points in the figure exhibit a clearpositive relationship (shown by the colored line) between labor productivity and exports. That is, those industries where the productivity of labor is relatively higher in the United States than in the United Kingdom are the industries with the higher ratios of U.S. to U.K. exports. This was true for the 20 industries shown in the figure (out of the total of 25 industries studied by MacDougall). The positive relationship between labor productivity and exports for the United States and the United Kingdom was confirmed by subsequent studies byBalassausing 1950 data andSternusing 1950 and 1959 data. Additional and more recent confirmation of the Ricardian trade model is provided by Golub (see Case Study 2-4).
2.7 Empirical Tests of the Ricardian Model 49
■CASE STUDY 2-4 Relative Unit Labor Costs and Relative Exports—United States and Japan In a 1995 study of the Ricardian trade model,
Golub examined relative unit labor costs (the ratio of wages to unit labor productivity) and the exports of the United States relative to those of the United Kingdom, Japan, Germany, Canada, and Australia and found that, in general, relative unit labor costs and exports were inversely related. That is, the higher the relative unit labor costs in the nation, the lower the relative exports of the nation, and vice versa. This relationship is particularly strong for U.S.-Japanese trade.
The colored line in Figure 2.5 shows a clear negative correlation between relative unit labor costs and relative exports for the 33 industries that
–0.2 0
–0.4 –0.6
–0.8 0.4 0.6 0.8 1
U.S./Japanese Unit Labor Costs
U.S./Japanese Exports 0.2
FIGURE 2.5. Relative Exports and Relative Unit Costs—United States and Japan.
The figure shows a clear negative correlation between relative exports and relative unit labor costs for 33 industries between the United States and Japan. It shows that the higher are U.S. relative labor costs, the lower are its exports in relation to Japan, thus supporting the Ricardian trade model.
Source: Adapted from S. S. Golub. Comparative and Absolute Advantage in the Asia-Pacific Region (San Francisco: Federal Reserve Bank of San Francisco, Center for Pacific Basin Monetary and Economic Studies, 1995). p. 46; and S. S. Golub and C. T.
Hsieh, ‘‘The Classical Ricardian Theory of Comparative Advantage Revisited,’’Review of International Economics, May 2000, pp. 221–234.
Golub studied for trade between the United States and Japan for 1990, thus lending additional support to the Ricardian trade model. Note that the relation- ship between relative unit labor costs and relative exports is negative in Figure 2.5, whereas the rela- tionship between relative unit labor productivities and exports shares is positive in Figure 2.4 because relative unit labor costs are the inverse of relative unit labor productivities. The above results were confirmed in a 2000 study by Golub and Hsieh for trade in the products of 39 sectors between the United States and nine other countries (Japan, Ger- many, the United Kingdom, France, Italy, Canada, Australia, Mexico, and Korea) from 1972 to 1991.
These empirical studies all seem to support the Ricardian theory of comparative advan- tage. That is, the actual pattern of trade seems to be based on the different labor productivities in different industries in the two nations. Production costs other than labor costs, demand considerations, political ties, and various obstructions to the flow of international trade did not break the link between relative labor productivity and export shares.
One possible question remained. Why did the United States not capture the entire export market from the United Kingdom (rather than only a rising share of exports) in those industries where it enjoyed a cost advantage (i.e., where the ratio of the productivity of U.S.
labor to U.K. labor was greater than 2)? MacDougall answered that this was due mainly to product differentiation. That is, the output of the same industry in the United States and the United Kingdom is not homogeneous. An American car is not identical to a British car.
Even if the American car is cheaper, some consumers in the rest of the world may still prefer the British car. Thus, the United Kingdom continues to export some cars even at a higher price. However, as the price difference grows, the United Kingdom’s share of car exports can be expected to decline. The same is true for most other products. Similarly, the United States continues to export to third markets some commodities in which it has a cost disadvantage with respect to the United Kingdom. We return to this important point in Section 6.4a.
Even though the simple Ricardian trade model has been empirically verified to a large extent, it has a serious shortcoming in that it assumes rather than explains comparative advantage. That is, Ricardo and classical economists in general provided no explanation for the difference in labor productivity and comparative advantage between nations, and they could not say much about the effect of international trade on the earnings of factors of production. By providing answers to both of these important questions, the Heckscher-Ohlin model (discussed in Chapter 5) theoretically improves on and extends the Ricardian model.
S U M M A R Y
1. This chapter examined the development of trade the- ory from the mercantilists to Smith, Ricardo, and Haberler and sought to answer two basic questions:
(a) What is the basis for and what are the gains from trade? and (b) What is the pattern of trade?
2. The mercantilists believed that a nation could gain in international trade only at the expense of other nations. As a result, they advocated restrictions on imports, incentives for exports, and strict government regulation of all economic activities.
3. According to Adam Smith, trade is based on absolute advantage and benefits both nations. (The discussion assumes a two-nation, two-commodity world.) That is, when each nation specializes in the production of the commodity of its absolute advantage and exchanges part of its output for the commodity of its absolute dis- advantage, both nations end up consuming more of both
commodities. Absolute advantage, however, explains only a small portion of international trade today.
4. David Ricardo introduced the law of comparative advantage. This postulates that even if one nation is less efficient than the other nation in the production of both commodities, there is still a basis for mutually beneficial trade (as long as the absolute disadvantage that the first nation has with respect to the second is not in the same proportion in both commodities). The less efficient nation should specialize in the produc- tion and export of the commodity in which its absolute disadvantage is smaller. (This is the commodity of its comparative advantage.) Ricardo, however, explained the law of comparative advantage in terms of the labor theory of value, which is unacceptable.
5. Gottfried Haberler came to the “rescue” by explain- ing the law of comparative advantage in terms of the
Questions for Review 51 opportunity cost theory. This states that the cost of a
commodity is the amount of a second commodity that must be given up to release just enough resources to produce one additional unit of the first commodity.
The opportunity cost of a commodity is equal to the relative price of that commodity and is given by the (absolute) slope of the production possibility frontier.
A straight-line production possibility frontier reflects constant opportunity costs.
6. In the absence of trade, a nation’s production pos- sibility frontier is also its consumption frontier. With trade, each nation can specialize in producing the com- modity of its comparative advantage and exchange part of its output with the other nation for the com- modity of its comparative disadvantage. By so doing, both nations end up consuming more of both com- modities than without trade. With complete specializa- tion, the equilibrium-relative commodity prices will be
between the pretrade-relative commodity prices pre- vailing in each nation.
7. The first empirical test of the Ricardian trade model was conducted by MacDougall in 1951 and 1952 using 1937 data. The results indicated that those indus- tries where labor productivity was relatively higher in the United States than in the United Kingdom were the industries with the higher ratios of U.S. to U.K.
exports to third markets. These results were confirmed by Balassa using 1950 data, Stern using 1950 and 1959 data, Golub using 1990 data, and Golub and Hsieh using 1972–1991 data. Thus, it can be seen that comparative advantage seems to be based on a differ- ence in labor productivity or costs, as postulated by Ricardo. However, the Ricardian model explains nei- ther the reason for the difference in labor productivity or costs across nations nor the effect of international trade on the earnings of factors.
A L O O K A H E A D
In Chapter 3, we examine the basis for and the gains from trade, as well as the pattern of trade in the more realistic case of increasing costs. Our model will then be completed in Chapter 4, where we see formally how the
rate at which commodities are exchanged in international trade is actually determined. This will also determine how the gains from trade are in fact divided between the two trading nations.
K E Y T E R M S Absolute advantage,
p. 34
Basis for trade, p. 31 Complete
specialization, p. 47
Constant opportunity costs, p. 43 Gains from trade,
p. 31 Labor theory of
value, p. 41
Laissez-faire, p. 35 Law of comparative
advantage, p. 36 Mercantilism, p. 32 Opportunity cost
theory, p. 42
Pattern of trade, p. 31 Production
possibility frontier, p. 42
Relative commodity prices, p. 44 Small-country case,
p. 47
Q U E S T I O N S F O R R E V I E W
1. What are the basic questions that we seek to answer in this chapter? In what way is the model presented in this chapter an abstraction or a simplification of the real world? Can the model be generalized?
2. What were the mercantilists’ views on trade? How does their concept of national wealth differ from today’s view?
3. Why is it important to study the mercantilists’
views on trade? How were their views different from those of Adam Smith? What is the relevance of all this today?
4. What was the basis for and the pattern of trade according to Adam Smith? How were gains from trade generated? What policies did Smith advocate
in international trade? What did he think was the proper function of government in the economic life of the nation?
5. In what way was Ricardo’s law of comparative advantage superior to Smith’s theory of absolute advantage? How do gains from trade arise with comparative advantage? How can a nation that is less efficient than another nation in the production of all commodities export anything to the second nation?
6. What is the exception to the law of comparative advantage? How prevalent is it?
7. Why is Ricardo’s explanation of the law of com- parative advantage unacceptable? What acceptable theory can be used to explain the law?
8. What is the relationship between opportunity costs and the production possibility frontier of a nation?
How does the production possibility frontier look under constant opportunity costs? What is the
relationship between the opportunity cost of a com- modity and the relative price of that commodity?
How can they be visualized graphically?
9. Why is a nation’s production possibility frontier the same as its consumption frontier in the absence of trade? How does the nation decide how much of each commodity to consume in the absence of trade?
10. What is meant by complete specialization? by incomplete specialization? Why do both nations gain from trade in the first instance but only the small nation in the second?
11. How is the combined supply curve of both nations for each of the traded commodities determined?
How is the equilibrium-relative commodity price determined with trade?
12. What are the results of empirical testing of the Ricardian model?
P R O B L E M S
1. Table 2.5 shows bushels of wheat and yards of cloth that the United States and the United King- dom can produce with one hour of labor time under four different hypothetical situations. In each case, identify the commodity in which the United States and the United Kingdom have an absolute advantage or disadvantage.
*2. With respect to Table 2.5, indicate in each case the commodity in which each nation has a com- parative advantage or disadvantage.
3. With respect to Table 2.5, indicate in each case whether or not trade is possible and the basis for trade.
*4. Suppose that in Case B in Table 2.5 the United States exchanges 4W for 4C with the United Kingdom.
■TABLE 2.5. Production Possibilities in the United States and the United Kingdom
Case A Case B Case C Case D
U.S. U.K. U.S. U.K. U.S. U.K. U.S. U.K.
Wheat (bushels/hour) 4 1 4 1 4 1 4 2
Cloth (yards/hour) 1 2 3 2 2 2 2 1
(a) How much does the United States gain in terms of cloth?
(b) How much does the United Kingdom gain in terms of cloth?
(c) What is the range for mutually beneficial trade?
(d) How much would each nation gain if they exchanged 4W for 6C instead?
5. Use the information for Case B in Table 2.5 and assume that labor is the only factor of production and is homogeneous (i.e., all of one type).
(a) What is the costin terms of labor content of producing wheat and cloth in the United States and the United Kingdom?
Appendix 53 (b) What is the dollar price of wheat and cloth
in the United States if the wage rate is $6?
(c) What is the pound price of wheat and cloth in the United Kingdom if the wage rate is £1?
6. Answer the following questions with reference to Problem 5.
(a) What is the dollar price of wheat and cloth in the United Kingdom if the exchange rate between the pound and the dollar is £1=$2? Would the United States be able to export wheat to the United Kingdom at this exchange rate? Would the United Kingdom be able to export cloth to the United States at this exchange rate?
(b) What if the exchange rate between the dollar and the pound were £1=$4?
(c) What if the exchange rate were £1=$1?
(d) What is the range of exchange rates that will allow the United States to export wheat to the United Kingdom and the United Kingdom to export cloth to the United States?
7. Assume that the data in Case B in Table 2.5 refer to millions of bushels of wheat and millions of yards of cloth.
(a) Plot on graph paper the production frontiers of the United States and the United Kingdom.
(b) What is the relative price of wheat (i.e., PW/PC) in the United States and in the United Kingdom in autarky (no trade)?
(c) What is the relative price of cloth (i.e., PC/PW) in the United States and in the United Kingdom in autarky?
8. Using the United States and United Kingdom production frontiers from Problem 7, assume that the no-trade or autarky point is 3W and 3/4C (in million units) in the United States and1/2W
and 1C in the United Kingdom. Also assume that with the opening of trade the United States exchanges 1W for 1C with the United Kingdom.
Show graphically for the United States and the United Kingdom the autarky (or no-trade) point of production and consumption, the point of pro- duction and consumption with trade, and the gains from trade.
9. (a)sfasfdWhat would be the equilibrium-relative com- modity price of wheat if DW(US+UK) shifted up by one-third in the left panel of Figure 2.3? How much wheat and cloth would the United States and the United Kingdom then produce?
(b) What does the answer to part (a) imply for DC(UK+US)in the right panel of Figure 2.3?
*10. What would happen ifDW(US+UK)intersected the horizontal portion of SW(US+UK) atPW/PC=2/3 and 120W in the left panel of Figure 2.3? What would this imply for specialization in produc- tion and the distribution in the gains from trade between the two nations?
11. Draw a figure similar to Figure 2.2 showing that the United Kingdom is now a small country, half the size shown in the right panel of Figure 2.2, and trades 20C for 30W with the United States at PW/PC=2/3.
12. (a)sfasfdHow was the Ricardian trade model tested empirically?
(b) In what way can the results be said to confirm the Ricardian model?
(c) Why do we then need other trade models?
13. How would you counter the argument that the United States needs to restrict textile imports in order to save American jobs?
*=Answer provided at www.wiley.com/college/
salvatore.
APPENDIX
We now extend the theory of comparative advantage first to the case of more than two commodities and then to the case of more than two nations. In each case, we will see that the theory of comparative advantage is easily generalized.